Sugerencias para otros estudios 107

1. Demographic variables

The demographics used in the study are Gender, Age, Educational status, Religion, ethnic background, Prior illness before the CVA, Lesion location and concordance.

2. CT scan to identify the area of lesion; however, this is normally done for every patient in the stroke unit therefore permission was only taken for its extraction from their files.

3. Physical disability:

The Barthel Index (BI) was first introduced by Mahoney and Barthel in 1965, which is now extensively used in rehabilitation. It was initially developed to

measure functional ability before and after treatment and to assess the amount of nursing care needed. Initially it was designed for use with long-stay hospitalized patients with neuromuscular or musculoskeletal problems Mahoney & Barthel (1965). It has subsequently been used and applied to evaluate treatment outcomes. It is extremely popular, and is one of the oldest and most widely used tests. The BI is based on a rating scale that is completed by an observer. It covers personal toileting, feeding, mobility from bed to chair, transfers, bathing, walking, dressing, incontinence and going upstairs. A total of 10 activities are scored, and the values are then added to give a total score ranging from 0 (totally dependent) to 100 (completely independent).

Lower scores indicate greater dependency. The BI measures what the patient actually does rather than what they can do. Information is obtained via verbal reports from patients, carers and staff, and by direct observation of some activities. The BI is sensitive, has concurrent and predictive validity, and is reliable (Mahoney & Barthel, 1965). It has good inter-rater, test–retest, and reported or observed reliability (Fricke & Unsworth, 1997). The internal consistency of the BI is extremely high, with a Cronbach‘s alpha coefficient of 0.98 Intra-observer and inter-observer reliabilities are high, with a Pearson‘s r score ranging from 0.89 to 0.99 (Shinar, Gross, Bronstein et al.,1987). In the present study it had a Cronbach alpha coefficient of 0.90 in the stroke population.

4. Health related locus of control:

The Multidimensional Health Locus of Control (MHLC) Scales were developed by Ken Wallston and colleagues. These scales were designed to

determined by the actions of individuals (as opposed to fate, luck, or chance) and, if so, whether the locus of that control is "internal" (i.e., residing in the person's own actions) or "external" (i.e., dependent on the actions of other people). The MHLC (Form C scales) was used, which separated externality into two dimensions powerful others and chance. The MHLC Scales had become the instrument of choice for health researchers wanting to assess perceived control of health. The three MHLC subscales are IHLC (eg, "The main thing that affects my health is what I myself do"), PHLC (e.g., "My family has a lot to do with my becoming sick or staying healthy"), and CHLC (e.g., "If it is meant to be, I will stay healthy"). In most populations, IHLC and PHLC are uncorrelated with each other, IHLC and CHLC are slightly negatively inter-correlated (-.l0 to -.20), and the two external dimensions, PHLC and CHLC, are somewhat positively inter-correlated (.20 to .30). The alpha reliabilities of the six-item subscales hover around .70 (.65-.75), and the test-retest reliabilities are in the range of .70-30 (Wallstron et.al, 2005).

Form C of the MHLC Scales designed to be a generic, medical-condition-specific assessment of locus of control belief. Each item of Form C contains the word "condition," which can be left intact or substituted with the name of an existing condition (e.g., "diabetes"). Form C has the same subscale structure as Forms NB, except that PHLC consists of two three-item subscales-"doctors" and "other people signifying a more complex discrimination of the role that physicians play in determining the health status of those already diagnosed. It consists of 18 items which had demonstrated excellent psychometric properties and is valid for routine use and clinical trials

in stroke population. In the stroke population of the study, it had a Cronbach‘s alpha coefficient of 0.65.

5. Becks depression inventory:

The original version of the Beck‘s Depression Inventory (Beck et al., 1961). It was published in 1961 and subsequently revised (Beck and Beck, 1972; Beck et al., 1979; Beck, 1988). It had been extensively used especially in assessing and monitoring changes with cognitive therapy and it was used to measure level of post-stroke depression. The long form of 21 items will be used to provide a quantitative assessment of the severity of depression. <10 represents minimum or no depression; 10-18 indicate mild to moderate; 19-29 showing moderate to severe depression; 30-63 is severe depression. Reliability studies showed a test-retest correlations having ranged from 0.48-0.90 (Beck et al., 1988), it had been extensively used in Nigeria with various validations; in the present study it had a Cronbach alpha coefficient of 0.77 in the stroke population.

6. The Hospital Anxiety and depression scale:

The Hospital Anxiety and Depression scale by Zigmond and Snaith (1983) was be used, it is a self-assessment scale that has been developed and found to be a reliable instrument for detecting states of depression and anxiety in the setting of a hospital medical outpatient clinic. The anxiety and depressive subscales are also valid measures of severity of the emotional disorder. The HADS comprises statements which the patient rates based on their experience over the past week. The 14 statements are relevant to generalised anxiety (7 statements) or 'depression' (again 7), the latter being largely (but not entirely)

take pleasure in everyday things enjoyed normally) HAD has good internal reliability (depression scale _ = 0.91, anxiety scale _ = .81, stress scale = .89).

Strong correlations were also found between scales with depression-anxiety r

=.42, anxiety-stress r =.46 and depression-stress r =.39. The scale had a Cronbach‘s Alpha of 0.64 in the stroke survivors.

7. Perceived social support:

The Perceived Multidimensional Social Support Scale (Zimet et al., 1988) is a twelve item scale of the level at which a patient feel he/she is been well integrated and cared for by those close to him/her. The psychometric properties of the Multidimensional Scale of Perceived Social Support have been demonstrated in diverse samples (Canty-Mitchell & Zimet, 2000). In the stroke population of the study, it had a Cronbach‘s alpha coefficient of 0.79.

8. Life events stress scale:

Life events were assessed (i.e., undesirable and severe). These events were also classified as primarily interpersonal (e.g.,death of a loved one) or related to achievement (e.g., loss of employment). The patients will be asked about events that occurred during the6-month period before the onset of the current post stroke -depressive episode.While also the comparison subjects would be asked about events during the 6-monthtime period immediately preceding the interview. The 6-monthtime frame was used because it has been shown to be the optimal time period for detecting an effect of life events on subsequent depressive onset, and other studies have used this standard, it had a 0.84 Cronbach Alpha reliability score.

9. Psychological index of stroke scale:

This scale is developed for use in the study and measures psychological distress in stroke patients. Items for the scale were drawn up from in-depth interviews with stroke survivors, where 22 items were initially produced. The items were further crosschecked for ambiguity or misunderstanding after which deleted items were replaced by new ones, then item total correlation was done while the non correlating items were removed. This was further crosschecked until all items appeared to be unproblematic. A response scale was decided upon using the Yes and No (1) or (0). Norms for the scale, reliability and item analysis was done to obtain a Cronbach‘s alpha of 0.60 after which it will be measured against the beck and hospital anxiety and depression instruments which are measure of psychological distress to determine its validity, it had a concurrent validity of 0.5 and 0.6 respectively.

Its content validity was further ascertained by factor analysis which showed good factor loading in a four-factor solution. This explained about 80% of its variance.

10. Stroke Levity Scale (SLS):

This was applied as an index of stroke severity. The SLS is a valid measure of stroke severity, which does not require the ability of the patient to read certain sentences written in the English language as this would exclude illiterate patients. The SLS also showed good internal consistency reliability with a Cronbach score of 0.70 and in the study 0.71